1/5 Oh wow. GPT-5.6-Sol seems to be on a whole other level. In just three weeks, it seems to have made nontrivial progress on 10+ problems in social choice. Exciting! (But I'm still processing what the existence of such a model means for research.) A snapshot:
2/5 We've started the slow part: hand-checking proofs (Fable 5 says they're correct), extracting insights, and writing human-readable papers. The first manuscript, with @ben_cookson_ and @_Paritosh_Verma, on (6) and (7) is out:
3/5 We're still verifying these results. So far, we haven't found any serious bugs in any of the proofs we've checked, including the full 41-page paper linked above. That level of reliability is materially different from all my testing of other publicly available models.
4/5 I'm building a website cataloguing social choice open problems, AI-discovered proofs, verification status, and human-readable expositions. Inspired by this beautiful work:
1/5 Oh wow. GPT-5.6-Sol seems to be on a whole other level. In just three weeks, it seems to have made nontrivial progress on 10+ problems in social choice. Exciting! (But I'm still processing what the existence of such a model means for research.) A snapshot:2/5 We've started the slow part: hand-checking proofs (Fable 5 says they're correct), extracting insights, and writing human-readable papers. The first manuscript, with @ben_cookson_ and @_Paritosh_Verma, on (6) and (7) is out:3/5 We're still verifying these results. So far, we haven't found any serious bugs in any of the proofs we've checked, including the full 41-page paper linked above. That level of reliability is materially different from all my testing of other publicly available models.4/5 I'm building a website cataloguing social choice open problems, AI-discovered proofs, verification status, and human-readable expositions. Inspired by this beautiful work:5/5 Hopefully, it won't take long, thanks to my dear friends Codex and Claude Code. Happy to join forces if you're building something similar.
@DominikPeters @polynoamial @ArielProcaccia @conitzer @harisazizk @hadi_hoss @BarmanSiddharth @ICaragiannis
yes
1/5 Oh wow. GPT-5.6-Sol seems to be on a whole other level. In just three weeks, it seems to have made nontrivial progress on 10+ problems in social choice. Exciting! (But I'm still processing what the existence of such a model means for research.) A snapshot: ... 2/5 We've started the slow part: hand-checking proofs (Fable 5 says they're correct), extracting insights, and writing human-readable papers. The first manuscript, with @ben_cookson_ and @_Paritosh_Verma, on (6) and (7) is out: ... 3/5 We're still verifying these results. So far, we haven't found any serious bugs in any of the proofs we've checked, including the full 41-page paper linked above. That level of reliability is materially different from all my testing of other publicly available models. ... 4/5 I'm building a website cataloguing social choice open problems, AI-discovered proofs, verification status, and human-readable expositions. Inspired by this beautiful work: ... 5/5 Hopefully, it won't take long, thanks to my dear friends Codex and Claude Code. Happy to join forces if you're building something similar.
@DominikPeters @polynoamial @ArielProcaccia @conitzer @harisazizk @hadi_hoss @BarmanSiddharth @ICaragiannis
Missing some Tweet in this thread? You can try to
Update
Unroll Another Thread
Convert any Twitter threads to an easy-to-read article instantly
Have you tried our Twitter bot?
You can now unroll any thread without leaving Twitter/X. Here's how to use our Twitter bot to do it.
